1. Field of the Invention
The invention relates to an image transforming method and, more particularly, to a method capable of automatically transforming a 2D image into a 3D image.
2. Description of the Prior Art
In general, a 3D image is formed by combining two sets of image data in different visual angles, wherein one set of image data corresponds to left-eye visual angle, and the other set of image data corresponds to right-eye visual angle. When a 3D image is displayed, the left eye of a viewer merely sees the image data corresponding to left-eye visual angle, and the right eye of the viewer merely sees the image data corresponding to right-eye visual angle. Accordingly, a stereoscopic vision is generated in the brain of the viewer.
In the prior art, when a user wants to transform a 2D image into a 3D image, at least one object in the 2D image needs to be selected. Afterward, the user himself judges whether the selected object belongs to a foreground or a background of the 2D image. Afterward, the user himself decides how much depth of field should be offered. If more than two objects are selected, the user needs to adjust each object respectively with the foregoing steps. It is very inconvenient for the user.
Therefore, scopes of the invention are to provide a method capable of automatically transforming a 2D image into a 3D image to solve the foregoing problems.